مقالههای a Heydarian
توجه: محتویات این صفحه به صورت خودکار پردازش شده و مقالههای نویسندگانی با تشابه اسمی، همگی در بخش یکسان نمایش داده میشوند.
اطلاعات انتشار: دومین کنفرانس بین المللی نفت، گاز و پتروشیمی، سال ۱۳۹۳
تعداد صفحات: ۹
In the context of oil production optimization, finding the well parameters that maximize an economical cost function such as net present value (NPV) is an important issue. Reservoir simulation in combination with automated optimization algorithms are often employed for this work. However, determining the optimal well design is a complex problem due to the reservoir heterogeneity, economic criteria and technical uncertainty. Therefore, it is necessary for the development of a powerful and trusted optimization algorithm that can detect best production variables with a minimum required number of simulation runs. This study presents a hybrid approach that employs genetic algorithm (GA) and general pattern search (GPS) to determine the optimum well locations in heterogeneous reservoir models. The hybrid algorithm entails some number of reservoir simulations using a GA method. The best solution found is then used as the initial guess for GPS. The overall algorithm takes advantage of the broad search provided by GA and the fast convergence to a local optimum provided by GPS. The performance of hybrid method has been compared with GA and GPS method. The hybrid method was found to generally outperform both standalone methods<\div>
نمایش نتایج ۱ تا ۱ از میان ۱ نتیجه